Xiaohua Liu

Title: Xiaohua Liu: Innovator in Natural Language Processing

Introduction

Xiaohua Liu is a prominent inventor based in Shenzhen, China. He has made significant contributions to the field of natural language processing, holding 2 patents that showcase his innovative approaches to sequence conversion methods.

Latest Patents

One of his latest patents is titled "Sequence conversion method and apparatus in natural language processing based on adjusting a weight associated with each word." This invention relates to natural language processing and discloses a method that includes obtaining a source sequence from an input signal and converting it into one or more source context vectors. The method further involves obtaining a target context vector corresponding to each source context vector and combining these to obtain the target sequence. A weight vector is applied to enhance the conversion process, improving the faithfulness of the source to target sequence conversion.

Another notable patent is "Sequence translation probability adjustment." This method involves receiving a source sequence and converting it into a source vector representation sequence. It includes obtaining at least two candidate target sequences and adjusting their translation probability values to select the most accurate output target sequence. This process enhances the loyalty of the target sequence to the source sequence during conversion.

Career Highlights

Xiaohua Liu has worked with notable companies such as Huawei Technologies Co., Limited and China Nuclear Power Design Company Ltd. in Shenzhen. His experience in these organizations has contributed to his expertise in the field of natural language processing.

Collaborations

Throughout his career, Xiaohua has collaborated with talented individuals, including Zhaopeng Tu and Hang Li. These collaborations have likely enriched his work and led to innovative advancements in his projects.

Conclusion

Xiaohua Liu is a distinguished inventor whose work in natural language processing has led to significant advancements in sequence conversion methods. His patents reflect his commitment to improving the accuracy and efficiency of language processing technologies.